An Investigation into the Use of Intelligent Systems for Currency Trading

Investors use a number of technical trading tools to help them in their decision-making. This article aims to enhance this decision making process through the application of Genetic Algorithms (GAs) and Artificial Neural Networks (ANNs). The signals generated by technical trading tools are optimised for maximum profit through the use of GAs. The optimised signals are fed into a variety of fully connected feed forward ANNs, which combine these signals and output a single set of signals of whether to buy, hold or sell in the current market state. The different solutions produced are compared and contrasted, to determine the best ANN architecture for this type of signal amalgamation problem, and the optimal population size and mutation function for the GA. The result is an autonomous trading system with intelligence. This system, as described in this article, has proven to be profitable based on data presented to it—which spans ten currencies over a five year period. The profit margins are statistically significant when compared to un-optimised trading rules as suggested by literature. Further, the margins are statistically significantly more profitable than other no-risk investment strategies.

[1]  Christopher J. Neely,et al.  Is Technical Analysis in the Foreign Exchange Market Profitable? A Genetic Programming Approach , 1996, Journal of Financial and Quantitative Analysis.

[2]  R. Sweeney,et al.  Beating the Foreign Exchange Market , 1986 .

[3]  E. Fama.  EFFICIENT CAPITAL MARKETS: A REVIEW OF THEORY AND EMPIRICAL WORK* , 1970 .

[4]  B. LeBaron,et al.  Simple Technical Trading Rules and the Stochastic Properties of Stock Returns , 1992 .

[5]  Richard M. Levich,et al.  The Merits of Active Currency Risk Management: Evidence from International Bond Portfolios , 1993 .

[6]  Spiros H. Martzoukos,et al.  Robust Artificial Neural Networks for Pricing of European Options , 2006 .

[7]  Mark P. Taylor,et al.  The use of technical analysis in the foreign exchange market , 1992 .

[8]  John Okunev,et al.  Do Momentum-Based Strategies Still Work in Foreign Currency Markets? , 2003, Journal of Financial and Quantitative Analysis.

[9]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[10]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

[11]  Cheolbeom Park,et al.  The Profitability of Technical Analysis: A Review , 2004 .

[12]  Tobias Hahn,et al.  Designing Short Term Trading Systems with Artificial Neural Networks , 2008, World Congress on Engineering.

[13]  Robert A. Levy,et al.  RELATIVE STRENGTH AS A CRITERION FOR INVESTMENT SELECTION , 1967 .

[14]  George Stephanides,et al.  Improving technical trading systems by using a new MATLAB-based genetic algorithm procedure , 2005, Math. Comput. Model..

[15]  Robert J. Marks,et al.  Neural Smithing: Supervised Learning in Feedforward Artificial Neural Networks , 1999 .

[16]  Khurshid M. Kiani,et al.  Testing Forecast Accuracy of Foreign Exchange Rates: Predictions from Feed Forward and Various Recurrent Neural Network Architectures , 2008 .

[17]  Melanie Mitchell,et al.  An introduction to genetic algorithms , 1996 .

[18]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .